INDEX

Explanations

references to buzzfeed and gawker

oai_token-act-pair · gpt-3.5-turbo

mentions of the media company BuzzFeed

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

semble

-0.84

atics

-0.75

 nonviolent

-0.69

oppable

-0.68

stood

-0.66

BuyableInstoreAndOnline

-0.65

jury

-0.65

abeth

-0.64

cludes

-0.64

arij

-0.63

POSITIVE LOGITS

Leaks

1.01

 BuzzFeed

0.95

 News

0.95

pedia

0.93

NEWS

0.91

Feed

0.87

Buzz

0.86

lus

0.84

lish

0.78

][

0.77

Activations Density 0.008%

references to buzzfeed and gawker

mentions of the media company BuzzFeed

No Comments

No Known Activations